ThoughtProbe: Classifier-Guided Thought Space Exploration Leveraging LLM Intrinsic Reasoning

• URL: http://arxiv.org/abs/2504.06650v1
• Date: Wed, 09 Apr 2025 07:37:27 GMT
• Title: ThoughtProbe: Classifier-Guided Thought Space Exploration Leveraging LLM Intrinsic Reasoning
• Authors: Zijian Wang, Chang Xu,
• Abstract summary: We make the key discovery that a simple linear classifier can effectively detect intrinsic reasoning capabilities in LLMs' activation space.<n>We propose a classifier-guided search framework that strategically explore a tree-structured response space.<n> Experimental results show that our framework's comprehensive exploration not only covers valid reasoning chains but also effectively identifies them.
• Score: 20.082244529609707
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Pre-trained large language models (LLMs) have been demonstrated to possess intrinsic reasoning capabilities that can emerge naturally when expanding the response space. However, the neural representation mechanisms underlying these intrinsic capabilities and approaches for their optimal utilization remain inadequately understood. In this work, we make the key discovery that a simple linear classifier can effectively detect intrinsic reasoning capabilities in LLMs' activation space, particularly within specific representation types and network layers. Based on this finding, we propose a classifier-guided search framework that strategically explore a tree-structured response space. In each node expansion, the classifier serves as a scoring and ranking mechanism that efficiently allocates computational resources by identifying and prioritizing more thoughtful reasoning directions for continuation. After completing the tree expansion, we collect answers from all branches to form a candidate answer pool. We propose a branch-aggregation selection method that marginalizes over all supporting branches by aggregating their thoughtfulness scores, thereby identifying the optimal answer from the pool. Experimental results show that our framework's comprehensive exploration not only covers valid reasoning chains but also effectively identifies them, achieving significant improvements across multiple arithmetic reasoning benchmarks.

Related papers

• Policy Guided Tree Search for Enhanced LLM Reasoning [3.090041654375235]
  Policy-Guided Tree Search (PGTS) is a framework that combines reinforcement learning with structured tree exploration to efficiently navigate reasoning paths.<n>Our key innovation is a learned policy that dynamically decides between expanding, branching, backtracking, or terminating exploration, eliminating the need for manuals or exhaustive search.
  arXiv  Detail & Related papers  (2025-02-04T22:08:20Z)
• AirRAG: Activating Intrinsic Reasoning for Retrieval Augmented Generation using Tree-based Search [4.4907551923591695]
  We propose a novel thinking pattern in RAG that integrates system analysis with efficient reasoning actions.<n>Specifically, our approach designs five fundamental reasoning actions, which are expanded to a broad tree-based reasoning space.<n> Experimental results demonstrate the effectiveness of AirRAG, showing significant performance gains on complex question-answering datasets.
  arXiv  Detail & Related papers  (2025-01-17T09:16:13Z)
• Enhancing LLM Reasoning with Reward-guided Tree Search [95.06503095273395]
  o1-like reasoning approach is challenging, and researchers have been making various attempts to advance this open area of research.<n>We present a preliminary exploration into enhancing the reasoning abilities of LLMs through reward-guided tree search algorithms.
  arXiv  Detail & Related papers  (2024-11-18T16:15:17Z)
• Optimized Feature Generation for Tabular Data via LLMs with Decision Tree Reasoning [53.241569810013836]
  We propose a novel framework that utilizes large language models (LLMs) to identify effective feature generation rules. We use decision trees to convey this reasoning information, as they can be easily represented in natural language. OCTree consistently enhances the performance of various prediction models across diverse benchmarks.
  arXiv  Detail & Related papers  (2024-06-12T08:31:34Z)
• Aggregation of Reasoning: A Hierarchical Framework for Enhancing Answer Selection in Large Language Models [84.15513004135576]
  Current research enhances the reasoning performance of Large Language Models (LLMs) by sampling multiple reasoning chains and ensembling based on the answer frequency. This approach fails in scenarios where the correct answers are in the minority. We introduce a hierarchical reasoning aggregation framework AoR, which selects answers based on the evaluation of reasoning chains.
  arXiv  Detail & Related papers  (2024-05-21T17:12:19Z)
• Logical Message Passing Networks with One-hop Inference on Atomic Formulas [57.47174363091452]
  We propose a framework for complex query answering that decomposes the Knowledge Graph embeddings from neural set operators. On top of the query graph, we propose the Logical Message Passing Neural Network (LMPNN) that connects the local one-hop inferences on atomic formulas to the global logical reasoning. Our approach yields the new state-of-the-art neural CQA model.
  arXiv  Detail & Related papers  (2023-01-21T02:34:06Z)
• Rationale-Augmented Ensembles in Language Models [53.45015291520658]
  We reconsider rationale-augmented prompting for few-shot in-context learning. We identify rationale sampling in the output space as the key component to robustly improve performance. We demonstrate that rationale-augmented ensembles achieve more accurate and interpretable results than existing prompting approaches.
  arXiv  Detail & Related papers  (2022-07-02T06:20:57Z)
• Optimal Counterfactual Explanations in Tree Ensembles [3.8073142980733]
  We advocate for a model-based search aiming at "optimal" explanations and propose efficient mixed-integer programming approaches. We show that isolation forests can be modeled within our framework to focus the search on plausible explanations with a low outlier score.
  arXiv  Detail & Related papers  (2021-06-11T22:44:27Z)
• Parameterizing Branch-and-Bound Search Trees to Learn Branching Policies [76.83991682238666]
  Branch and Bound (B&B) is the exact tree search method typically used to solve Mixed-Integer Linear Programming problems (MILPs) We propose a novel imitation learning framework, and introduce new input features and architectures to represent branching.
  arXiv  Detail & Related papers  (2020-02-12T17:43:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.